diff options
Diffstat (limited to 'fs/btrfs')
| -rw-r--r-- | fs/btrfs/backref.c | 6 | ||||
| -rw-r--r-- | fs/btrfs/backref.h | 3 | ||||
| -rw-r--r-- | fs/btrfs/block-group.c | 367 | ||||
| -rw-r--r-- | fs/btrfs/block-group.h | 6 | ||||
| -rw-r--r-- | fs/btrfs/compression.c | 2 | ||||
| -rw-r--r-- | fs/btrfs/ctree.c | 67 | ||||
| -rw-r--r-- | fs/btrfs/delayed-ref.c | 4 | ||||
| -rw-r--r-- | fs/btrfs/disk-io.c | 2 | ||||
| -rw-r--r-- | fs/btrfs/extent-tree.c | 3 | ||||
| -rw-r--r-- | fs/btrfs/inode.c | 159 | ||||
| -rw-r--r-- | fs/btrfs/ordered-data.c | 2 | ||||
| -rw-r--r-- | fs/btrfs/ordered-data.h | 3 | ||||
| -rw-r--r-- | fs/btrfs/qgroup.c | 38 | ||||
| -rw-r--r-- | fs/btrfs/qgroup.h | 2 | ||||
| -rw-r--r-- | fs/btrfs/tests/qgroup-tests.c | 20 | ||||
| -rw-r--r-- | fs/btrfs/transaction.c | 15 | ||||
| -rw-r--r-- | fs/btrfs/transaction.h | 9 | ||||
| -rw-r--r-- | fs/btrfs/tree-log.c | 37 | ||||
| -rw-r--r-- | fs/btrfs/volumes.c | 356 | ||||
| -rw-r--r-- | fs/btrfs/volumes.h | 5 | ||||
| -rw-r--r-- | fs/btrfs/zoned.c | 12 |
21 files changed, 779 insertions, 339 deletions
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 7a8a2fc19533..78b202d198b8 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -1488,15 +1488,15 @@ static int btrfs_find_all_roots_safe(struct btrfs_trans_handle *trans, int btrfs_find_all_roots(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info, u64 bytenr, u64 time_seq, struct ulist **roots, - bool ignore_offset) + bool ignore_offset, bool skip_commit_root_sem) { int ret; - if (!trans) + if (!trans && !skip_commit_root_sem) down_read(&fs_info->commit_root_sem); ret = btrfs_find_all_roots_safe(trans, fs_info, bytenr, time_seq, roots, ignore_offset); - if (!trans) + if (!trans && !skip_commit_root_sem) up_read(&fs_info->commit_root_sem); return ret; } diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h index 17abde7f794c..ff5f07f9940b 100644 --- a/fs/btrfs/backref.h +++ b/fs/btrfs/backref.h @@ -47,7 +47,8 @@ int btrfs_find_all_leafs(struct btrfs_trans_handle *trans, const u64 *extent_item_pos, bool ignore_offset); int btrfs_find_all_roots(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info, u64 bytenr, - u64 time_seq, struct ulist **roots, bool ignore_offset); + u64 time_seq, struct ulist **roots, bool ignore_offset, + bool skip_commit_root_sem); char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path, u32 name_len, unsigned long name_off, struct extent_buffer *eb_in, u64 parent, diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 38b127b9edfc..9e7d9d0c763d 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -1498,9 +1498,18 @@ void btrfs_reclaim_bgs_work(struct work_struct *work) if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_BALANCE)) return; - mutex_lock(&fs_info->reclaim_bgs_lock); + /* + * Long running balances can keep us blocked here for eternity, so + * simply skip reclaim if we're unable to get the mutex. + */ + if (!mutex_trylock(&fs_info->reclaim_bgs_lock)) { + btrfs_exclop_finish(fs_info); + return; + } + spin_lock(&fs_info->unused_bgs_lock); while (!list_empty(&fs_info->reclaim_bgs)) { + u64 zone_unusable; int ret = 0; bg = list_first_entry(&fs_info->reclaim_bgs, @@ -1534,13 +1543,22 @@ void btrfs_reclaim_bgs_work(struct work_struct *work) goto next; } + /* + * Cache the zone_unusable value before turning the block group + * to read only. As soon as the blog group is read only it's + * zone_unusable value gets moved to the block group's read-only + * bytes and isn't available for calculations anymore. + */ + zone_unusable = bg->zone_unusable; ret = inc_block_group_ro(bg, 0); up_write(&space_info->groups_sem); if (ret < 0) goto next; - btrfs_info(fs_info, "reclaiming chunk %llu with %llu%% used", - bg->start, div_u64(bg->used * 100, bg->length)); + btrfs_info(fs_info, + "reclaiming chunk %llu with %llu%% used %llu%% unusable", + bg->start, div_u64(bg->used * 100, bg->length), + div64_u64(zone_unusable * 100, bg->length)); trace_btrfs_reclaim_block_group(bg); ret = btrfs_relocate_chunk(fs_info, bg->start); if (ret) @@ -2197,6 +2215,13 @@ error: return ret; } +/* + * This function, insert_block_group_item(), belongs to the phase 2 of chunk + * allocation. + * + * See the comment at btrfs_chunk_alloc() for details about the chunk allocation + * phases. + */ static int insert_block_group_item(struct btrfs_trans_handle *trans, struct btrfs_block_group *block_group) { @@ -2219,15 +2244,19 @@ static int insert_block_group_item(struct btrfs_trans_handle *trans, return btrfs_insert_item(trans, root, &key, &bgi, sizeof(bgi)); } +/* + * This function, btrfs_create_pending_block_groups(), belongs to the phase 2 of + * chunk allocation. + * + * See the comment at btrfs_chunk_alloc() for details about the chunk allocation + * phases. + */ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans) { struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_block_group *block_group; int ret = 0; - if (!trans->can_flush_pending_bgs) - return; - while (!list_empty(&trans->new_bgs)) { int index; @@ -2242,6 +2271,13 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans) ret = insert_block_group_item(trans, block_group); if (ret) btrfs_abort_transaction(trans, ret); + if (!block_group->chunk_item_inserted) { + mutex_lock(&fs_info->chunk_mutex); + ret = btrfs_chunk_alloc_add_chunk_item(trans, block_group); + mutex_unlock(&fs_info->chunk_mutex); + if (ret) + btrfs_abort_transaction(trans, ret); + } ret = btrfs_finish_chunk_alloc(trans, block_group->start, block_group->length); if (ret) @@ -2265,8 +2301,9 @@ next: btrfs_trans_release_chunk_metadata(trans); } -int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used, - u64 type, u64 chunk_offset, u64 size) +struct btrfs_block_group *btrfs_make_block_group(struct btrfs_trans_handle *trans, + u64 bytes_used, u64 type, + u64 chunk_offset, u64 size) { struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_block_group *cache; @@ -2276,7 +2313,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used, cache = btrfs_create_block_group_cache(fs_info, chunk_offset); if (!cache) - return -ENOMEM; + return ERR_PTR(-ENOMEM); cache->length = size; set_free_space_tree_thresholds(cache); @@ -2290,7 +2327,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used, ret = btrfs_load_block_group_zone_info(cache, true); if (ret) { btrfs_put_block_group(cache); - return ret; + return ERR_PTR(ret); } ret = exclude_super_stripes(cache); @@ -2298,7 +2335,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used, /* We may have excluded something, so call this just in case */ btrfs_free_excluded_extents(cache); btrfs_put_block_group(cache); - return ret; + return ERR_PTR(ret); } add_new_free_space(cache, chunk_offset, chunk_offset + size); @@ -2325,7 +2362,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used, if (ret) { btrfs_remove_free_space_cache(cache); btrfs_put_block_group(cache); - return ret; + return ERR_PTR(ret); } /* @@ -2344,7 +2381,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used, btrfs_update_delayed_refs_rsv(trans); set_avail_alloc_bits(fs_info, type); - return 0; + return cache; } /* @@ -3222,11 +3259,203 @@ int btrfs_force_chunk_alloc(struct btrfs_trans_handle *trans, u64 type) return btrfs_chunk_alloc(trans, alloc_flags, CHUNK_ALLOC_FORCE); } +static int do_chunk_alloc(struct btrfs_trans_handle *trans, u64 flags) +{ + struct btrfs_block_group *bg; + int ret; + + /* + * Check if we have enough space in the system space info because we + * will need to update device items in the chunk btree and insert a new + * chunk item in the chunk btree as well. This will allocate a new + * system block group if needed. + */ + check_system_chunk(trans, flags); + + bg = btrfs_alloc_chunk(trans, flags); + if (IS_ERR(bg)) { + ret = PTR_ERR(bg); + goto out; + } + + /* + * If this is a system chunk allocation then stop right here and do not + * add the chunk item to the chunk btree. This is to prevent a deadlock + * because this system chunk allocation can be triggered while COWing + * some extent buffer of the chunk btree and while holding a lock on a + * parent extent buffer, in which case attempting to insert the chunk + * item (or update the device item) would result in a deadlock on that + * parent extent buffer. In this case defer the chunk btree updates to + * the second phase of chunk allocation and keep our reservation until + * the second phase completes. + * + * This is a rare case and can only be triggered by the very few cases + * we have where we need to touch the chunk btree outside chunk allocation + * and chunk removal. These cases are basically adding a device, removing + * a device or resizing a device. + */ + if (flags & BTRFS_BLOCK_GROUP_SYSTEM) + return 0; + + ret = btrfs_chunk_alloc_add_chunk_item(trans, bg); + /* + * Normally we are not expected to fail with -ENOSPC here, since we have + * previously reserved space in the system space_info and allocated one + * new system chunk if necessary. However there are two exceptions: + * + * 1) We may have enough free space in the system space_info but all the + * existing system block groups have a profile which can not be used + * for extent allocation. + * + * This happens when mounting in degraded mode. For example we have a + * RAID1 filesystem with 2 devices, lose one device and mount the fs + * using the other device in degraded mode. If we then allocate a chunk, + * we may have enough free space in the existing system space_info, but + * none of the block groups can be used for extent allocation since they + * have a RAID1 profile, and because we are in degraded mode with a + * single device, we are forced to allocate a new system chunk with a + * SINGLE profile. Making check_system_chunk() iterate over all system + * block groups and check if they have a usable profile and enough space + * can be slow on very large filesystems, so we tolerate the -ENOSPC and + * try again after forcing allocation of a new system chunk. Like this + * we avoid paying the cost of that search in normal circumstances, when + * we were not mounted in degraded mode; + * + * 2) We had enough free space info the system space_info, and one suitable + * block group to allocate from when we called check_system_chunk() + * above. However right after we called it, the only system block group + * with enough free space got turned into RO mode by a running scrub, + * and in this case we have to allocate a new one and retry. We only + * need do this allocate and retry once, since we have a transaction + * handle and scrub uses the commit root to search for block groups. + */ + if (ret == -ENOSPC) { + const u64 sys_flags = btrfs_system_alloc_profile(trans->fs_info); + struct btrfs_block_group *sys_bg; + + sys_bg = btrfs_alloc_chunk(trans, sys_flags); + if (IS_ERR(sys_bg)) { + ret = PTR_ERR(sys_bg); + btrfs_abort_transaction(trans, ret); + goto out; + } + + ret = btrfs_chunk_alloc_add_chunk_item(trans, sys_bg); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out; + } + + ret = btrfs_chunk_alloc_add_chunk_item(trans, bg); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out; + } + } else if (ret) { + btrfs_abort_transaction(trans, ret); + goto out; + } +out: + btrfs_trans_release_chunk_metadata(trans); + + return ret; +} + /* - * If force is CHUNK_ALLOC_FORCE: + * Chunk allocation is done in 2 phases: + * + * 1) Phase 1 - through btrfs_chunk_alloc() we allocate device extents for + * the chunk, the chunk mapping, create its block group and add the items + * that belong in the chunk btree to it - more specifically, we need to + * update device items in the chunk btree and add a new chunk item to it. + * + * 2) Phase 2 - through btrfs_create_pending_block_groups(), we add the block + * group item to the extent btree and the device extent items to the devices + * btree. + * + * This is done to prevent deadlocks. For example when COWing a node from the + * extent btree we are holding a write lock on the node's parent and if we + * trigger chunk allocation and attempted to insert the new block group item + * in the extent btree right way, we could deadlock because the path for the + * insertion can include that parent node. At first glance it seems impossible + * to trigger chunk allocation after starting a transaction since tasks should + * reserve enough transaction units (metadata space), however while that is true + * most of the time, chunk allocation may still be triggered for several reasons: + * + * 1) When reserving metadata, we check if there is enough free space in the + * metadata space_info and therefore don't trigger allocation of a new chunk. + * However later when the task actually tries to COW an extent buffer from + * the extent btree or from the device btree for example, it is forced to + * allocate a new block group (chunk) because the only one that had enough + * free space was just turned to RO mode by a running scrub for example (or + * device replace, block group reclaim thread, etc), so we can not use it + * for allocating an extent and end up being forced to allocate a new one; + * + * 2) Because we only check that the metadata space_info has enough free bytes, + * we end up not allocating a new metadata chunk in that case. However if + * the filesystem was mounted in degraded mode, none of the existing block + * groups might be suitable for extent allocation due to their incompatible + * profile (for e.g. mounting a 2 devices filesystem, where all block groups + * use a RAID1 profile, in degraded mode using a single device). In this case + * when the task attempts to COW some extent buffer of the extent btree for + * example, it will trigger allocation of a new metadata block group with a + * suitable profile (SINGLE profile in the example of the degraded mount of + * the RAID1 filesystem); + * + * 3) The task has reserved enough transaction units / metadata space, but when + * it attempts to COW an extent buffer from the extent or device btree for + * example, it does not find any free extent in any metadata block group, + * therefore forced to try to allocate a new metadata block group. + * This is because some other task allocated all available extents in the + * meanwhile - this typically happens with tasks that don't reserve space + * properly, either intentionally or as a bug. One example where this is + * done intentionally is fsync, as it does not reserve any transaction units + * and ends up allocating a variable number of metadata extents for log + * tree extent buffers. + * + * We also need this 2 phases setup when adding a device to a filesystem with + * a seed device - we must create new metadata and system chunks without adding + * any of the block group items to the chunk, extent and device btrees. If we + * did not do it this way, we would get ENOSPC when attempting to update those + * btrees, since all the chunks from the seed device are read-only. + * + * Phase 1 does the updates and insertions to the chunk btree because if we had + * it done in phase 2 and have a thundering herd of tasks allocating chunks in + * parallel, we risk having too many system chunks allocated by many tasks if + * many tasks reach phase 1 without the previous ones completing phase 2. In the + * extreme case this leads to exhaustion of the system chunk array in the + * superblock. This is easier to trigger if using a btree node/leaf size of 64K + * and with RAID filesystems (so we have more device items in the chunk btree). + * This has happened before and commit eafa4fd0ad0607 ("btrfs: fix exhaustion of + * the system chunk array due to concurrent allocations") provides more details. + * + * For allocation of system chunks, we defer the updates and insertions into the + * chunk btree to phase 2. This is to prevent deadlocks on extent buffers because + * if the chunk allocation is triggered while COWing an extent buffer of the + * chunk btree, we are holding a lock on the parent of that extent buffer and + * doing the chunk btree updates and insertions can require locking that parent. + * This is for the very few and rare cases where we update the chunk btree that + * are not chunk allocation or chunk removal: adding a device, removing a device + * or resizing a device. + * + * The reservation of system space, done through check_system_chunk(), as well + * as all the updates and insertions into the chunk btree must be done while + * holding fs_info->chunk_mutex. This is important to guarantee that while COWing + * an extent buffer from the chunks btree we never trigger allocation of a new + * system chunk, which would result in a deadlock (trying to lock twice an + * extent buffer of the chunk btree, first time before triggering the chunk + * allocation and the second time during chunk allocation while attempting to + * update the chunks btree). The system chunk array is also updated while holding + * that mutex. The same logic applies to removing chunks - we must reserve system + * space, update the chunk btree and the system chunk array in the superblock + * while holding fs_info->chunk_mutex. + * + * This function, btrfs_chunk_alloc(), belongs to phase 1. + * + * If @force is CHUNK_ALLOC_FORCE: * - return 1 if it successfully allocates a chunk, * - return errors including -ENOSPC otherwise. - * If force is NOT CHUNK_ALLOC_FORCE: + * If @force is NOT CHUNK_ALLOC_FORCE: * - return 0 if it doesn't need to allocate a new chunk, * - return 1 if it successfully allocates a chunk, * - return errors including -ENOSPC otherwise. @@ -3243,6 +3472,13 @@ int btrfs_chunk_alloc(struct btrfs_trans_handle *trans, u64 flags, /* Don't re-enter if we're already allocating a chunk */ if (trans->allocating_chunk) return -ENOSPC; + /* + * If we are removing a chunk, don't re-enter or we would deadlock. + * System space reservation and system chunk allocation is done by the + * chunk remove operation (btrfs_remove_chunk()). + */ + if (trans->removing_chunk) + return -ENOSPC; space_info = btrfs_find_space_info(fs_info, flags); ASSERT(space_info); @@ -3306,13 +3542,7 @@ int btrfs_chunk_alloc(struct btrfs_trans_handle *trans, u64 flags, force_metadata_allocation(fs_info); } - /* - * Check if we have enough space in SYSTEM chunk because we may need - * to update devices. - */ - check_system_chunk(trans, flags); - - ret = btrfs_alloc_chunk(trans, flags); + ret = do_chunk_alloc(trans, flags); trans->allocating_chunk = false; spin_lock(&space_info->lock); @@ -3331,22 +3561,6 @@ out: space_info->chunk_alloc = 0; spin_unlock(&space_info->lock); mutex_unlock(&fs_info->chunk_mutex); - /* - * When we allocate a new chunk we reserve space in the chunk block - * reserve to make sure we can COW nodes/leafs in the chunk tree or - * add new nodes/leafs to it if we end up needing to do it when - * inserting the chunk item and updating device items as part of the - * second phase of chunk allocation, performed by - * btrfs_finish_chunk_alloc(). So make sure we don't accumulate a - * large number of new block groups to create in our transaction - * handle's new_bgs list to avoid exhausting the chunk block reserve - * in extreme cases - like having a single transaction create many new - * block groups when starting to write out the free space caches of all - * the block groups that were made dirty during the lifetime of the - * transaction. - */ - if (trans->chunk_bytes_reserved >= (u64)SZ_2M) - btrfs_create_pending_block_groups(trans); return ret; } @@ -3367,7 +3581,6 @@ static u64 get_profile_num_devs(struct btrfs_fs_info *fs_info, u64 type) */ void check_system_chunk(struct btrfs_trans_handle *trans, u64 type) { - struct btrfs_transaction *cur_trans = trans->transaction; struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_space_info *info; u64 left; @@ -3382,7 +3595,6 @@ void check_system_chunk(struct btrfs_trans_handle *trans, u64 type) lockdep_assert_held(&fs_info->chunk_mutex); info = btrfs_find_space_info(fs_info, BTRFS_BLOCK_GROUP_SYSTEM); -again: spin_lock(&info->lock); left = info->total_bytes - btrfs_space_info_used(info, true); spin_unlock(&info->lock); @@ -3401,76 +3613,39 @@ again: if (left < thresh) { u64 flags = btrfs_system_alloc_profile(fs_info); - u64 reserved = atomic64_read(&cur_trans->chunk_bytes_reserved); - - /* - * If there's not available space for the chunk tree (system - * space) and there are other tasks that reserved space for - * creating a new system block group, wait for them to complete - * the creation of their system block group and release excess - * reserved space. We do this because: - * - * *) We can end up allocating more system chunks than necessary - * when there are multiple tasks that are concurrently - * allocating block groups, which can lead to exhaustion of - * the system array in the superblock; - * - * *) If we allocate extra and unnecessary system block groups, - * despite being empty for a long time, and possibly forever, - * they end not being added to the list of unused block groups - * because that typically happens only when deallocating the - * last extent from a block group - which never happens since - * we never allocate from them in the first place. The few - * exceptions are when mounting a filesystem or running scrub, - * which add unused block groups to the list of unused block - * groups, to be deleted by the cleaner kthread. - * And even when they are added to the list of unused block - * groups, it can take a long time until they get deleted, - * since the cleaner kthread might be sleeping or busy with - * other work (deleting subvolumes, running delayed iputs, - * defrag scheduling, etc); - * - * This is rare in practice, but can happen when too many tasks - * are allocating blocks groups in parallel (via fallocate()) - * and before the one that reserved space for a new system block - * group finishes the block group creation and releases the space - * reserved in excess (at btrfs_create_pending_block_groups()), - * other tasks end up here and see free system space temporarily - * not enough for updating the chunk tree. - * - * We unlock the chunk mutex before waiting for such tasks and - * lock it again after the wait, otherwise we would deadlock. - * It is safe to do so because allocating a system chunk is the - * first thing done while allocating a new block group. - */ - if (reserved > trans->chunk_bytes_reserved) { - const u64 min_needed = reserved - thresh; - - mutex_unlock(&fs_info->chunk_mutex); - wait_event(cur_trans->chunk_reserve_wait, - atomic64_read(&cur_trans->chunk_bytes_reserved) <= - min_needed); - mutex_lock(&fs_info->chunk_mutex); - goto again; - } + struct btrfs_block_group *bg; /* * Ignore failure to create system chunk. We might end up not * needing it, as we might not need to COW all nodes/leafs from * the paths we visit in the chunk tree (they were already COWed * or created in the current transaction for example). + * + * Also, if our caller is allocating a system chunk, do not + * attempt to insert the chunk item in the chunk btree, as we + * could deadlock on an extent buffer since our caller may be + * COWing an extent buffer from the chunk btree. */ - ret = btrfs_alloc_chunk(trans, flags); + bg = btrfs_alloc_chunk(trans, flags); + if (IS_ERR(bg)) { + ret = PTR_ERR(bg); + } else if (!(type & BTRFS_BLOCK_GROUP_SYSTEM)) { + /* + * If we fail to add the chunk item here, we end up + * trying again at phase 2 of chunk allocation, at + * btrfs_create_pending_block_groups(). So ignore + * any error here. + */ + btrfs_chunk_alloc_add_chunk_item(trans, bg); + } } if (!ret) { ret = btrfs_block_rsv_add(fs_info->chunk_root, &fs_info->chunk_block_rsv, thresh, BTRFS_RESERVE_NO_FLUSH); - if (!ret) { - atomic64_add(thresh, &cur_trans->chunk_bytes_reserved); + if (!ret) trans->chunk_bytes_reserved += thresh; - } } } diff --git a/fs/btrfs/block-group.h b/fs/btrfs/block-group.h index 7b927425dc71..c72a71efcb18 100644 --- a/fs/btrfs/block-group.h +++ b/fs/btrfs/block-group.h @@ -97,6 +97,7 @@ struct btrfs_block_group { unsigned int removed:1; unsigned int to_copy:1; unsigned int relocating_repair:1; + unsigned int chunk_item_inserted:1; int disk_cache_state; @@ -268,8 +269,9 @@ void btrfs_reclaim_bgs_work(struct work_struct *work); void btrfs_reclaim_bgs(struct btrfs_fs_info *fs_info); void btrfs_mark_bg_to_reclaim(struct btrfs_block_group *bg); int btrfs_read_block_groups(struct btrfs_fs_info *info); -int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used, - u64 type, u64 chunk_offset, u64 size); +struct btrfs_block_group *btrfs_make_block_group(struct btrfs_trans_handle *trans, + u64 bytes_used, u64 type, + u64 chunk_offset, u64 size); void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans); int btrfs_inc_block_group_ro(struct btrfs_block_group *cache, bool do_chunk_alloc); diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index 9a023ae0f98b..30d82cdf128c 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -352,7 +352,7 @@ static void end_compressed_bio_write(struct bio *bio) btrfs_record_physical_zoned(inode, cb->start, bio); btrfs_writepage_endio_finish_ordered(BTRFS_I(inode), NULL, cb->start, cb->start + cb->len - 1, - bio->bi_status == BLK_STS_OK); + !cb->errors); end_compressed_writeback(inode, cb); /* note, our inode could be gone now */ diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index 4bc3ca2cbd7d..c5c08c87e130 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -364,49 +364,6 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans, return 0; } -static struct extent_buffer *alloc_tree_block_no_bg_flush( - struct btrfs_trans_handle *trans, - struct btrfs_root *root, - u64 parent_start, - const struct btrfs_disk_key *disk_key, - int level, - u64 hint, - u64 empty_size, - enum btrfs_lock_nesting nest) -{ - struct btrfs_fs_info *fs_info = root->fs_info; - struct extent_buffer *ret; - - /* - * If we are COWing a node/leaf from the extent, chunk, device or free - * space trees, make sure that we do not finish block group creation of - * pending block groups. We do this to avoid a deadlock. - * COWing can result in allocation of a new chunk, and flushing pending - * block groups (btrfs_create_pending_block_groups()) can be triggered - * when finishing allocation of a new chunk. Creation of a pending block - * group modifies the extent, chunk, device and free space trees, - * therefore we could deadlock with ourselves since we are holding a - * lock on an extent buffer that btrfs_create_pending_block_groups() may - * try to COW later. - * For similar reasons, we also need to delay flushing pending block - * groups when splitting a leaf or node, from one of those trees, since - * we are holding a write lock on it and its parent or when inserting a - * new root node for one of those trees. - */ - if (root == fs_info->extent_root || - root == fs_info->chunk_root || - root == fs_info->dev_root || - root == fs_info->free_space_root) - trans->can_flush_pending_bgs = false; - - ret = btrfs_alloc_tree_block(trans, root, parent_start, - root->root_key.objectid, disk_key, level, - hint, empty_size, nest); - trans->can_flush_pending_bgs = true; - - return ret; -} - /* * does the dirty work in cow of a single block. The parent block (if * supplied) is updated to point to the new cow copy. The new buffer is marked @@ -455,8 +412,9 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans, if ((root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) && parent) parent_start = parent->start; - cow = alloc_tree_block_no_bg_flush(trans, root, parent_start, &disk_key, - level, search_start, empty_size, nest); + cow = btrfs_alloc_tree_block(trans, root, parent_start, + root->root_key.objectid, &disk_key, level, + search_start, empty_size, nest); if (IS_ERR(cow)) return PTR_ERR(cow); @@ -2458,9 +2416,9 @@ static noinline int insert_new_root(struct btrfs_trans_handle *trans, else btrfs_node_key(lower, &lower_key, 0); - c = alloc_tree_block_no_bg_flush(trans, root, 0, &lower_key, level, - root->node->start, 0, - BTRFS_NESTING_NEW_ROOT); + c = btrfs_alloc_tree_block(trans, root, 0, root->root_key.objectid, + &lower_key, level, root->node->start, 0, + BTRFS_NESTING_NEW_ROOT); if (IS_ERR(c)) return PTR_ERR(c); @@ -2589,8 +2547,9 @@ static noinline int split_node(struct btrfs_trans_handle *trans, mid = (c_nritems + 1) / 2; btrfs_node_key(c, &disk_key, mid); - split = alloc_tree_block_no_bg_flush(trans, root, 0, &disk_key, level, - c->start, 0, BTRFS_NESTING_SPLIT); + split = btrfs_alloc_tree_block(trans, root, 0, root->root_key.objectid, + &disk_key, level, c->start, 0, + BTRFS_NESTING_SPLIT); if (IS_ERR(split)) return PTR_ERR(split); @@ -3381,10 +3340,10 @@ again: * BTRFS_NESTING_SPLIT_THE_SPLITTENING if we need to, but for now just * use BTRFS_NESTING_NEW_ROOT. */ - right = alloc_tree_block_no_bg_flush(trans, root, 0, &disk_key, 0, - l->start, 0, num_doubles ? - BTRFS_NESTING_NEW_ROOT : - BTRFS_NESTING_SPLIT); + right = btrfs_alloc_tree_block(trans, root, 0, root->root_key.objectid, + &disk_key, 0, l->start, 0, + num_doubles ? BTRFS_NESTING_NEW_ROOT : + BTRFS_NESTING_SPLIT); if (IS_ERR(right)) return PTR_ERR(right); diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 06bc842ecdb3..ca848b183474 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -974,7 +974,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref); if (qrecord_inserted) - btrfs_qgroup_trace_extent_post(fs_info, record); + btrfs_qgroup_trace_extent_post(trans, record); return 0; } @@ -1069,7 +1069,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, if (qrecord_inserted) - return btrfs_qgroup_trace_extent_post(fs_info, record); + return btrfs_qgroup_trace_extent_post(trans, record); return 0; } diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index b117dd3b8172..a59ab7b9aea0 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -209,7 +209,7 @@ void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb, static void csum_tree_block(struct extent_buffer *buf, u8 *result) { struct btrfs_fs_info *fs_info = buf->fs_info; - const int num_pages = fs_info->nodesize >> PAGE_SHIFT; + const int num_pages = num_extent_pages(buf); const int first_page_part = min_t(u32, PAGE_SIZE, fs_info->nodesize); SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); char *kaddr; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index d296483d148f..268ce58d4569 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -6019,6 +6019,9 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range) mutex_lock(&fs_info->fs_devices->device_list_mutex); devices = &fs_info->fs_devices->devices; list_for_each_entry(device, devices, dev_list) { + if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state)) + continue; + ret = btrfs_trim_free_extents(device, &group_trimmed); if (ret) { dev_failed++; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index e6eb20987351..06f9f167222b 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2271,13 +2271,127 @@ static blk_status_t btrfs_submit_bio_start(struct inode *inode, struct bio *bio, return btrfs_csum_one_bio(BTRFS_I(inode), bio, 0, 0); } +/* + * Split an extent_map at [start, start + len] + * + * This function is intended to be used only for extract_ordered_extent(). + */ +static int split_zoned_em(struct btrfs_inode *inode, u64 start, u64 len, + u64 pre, u64 post) +{ + struct extent_map_tree *em_tree = &inode->extent_tree; + struct extent_map *em; + struct extent_map *split_pre = NULL; + struct extent_map *split_mid = NULL; + struct extent_map *split_post = NULL; + int ret = 0; + int modified; + unsigned long flags; + + /* Sanity check */ + if (pre == 0 && post == 0) + return 0; + + split_pre = alloc_extent_map(); + if (pre) + split_mid = alloc_extent_map(); + if (post) + split_post = alloc_extent_map(); + if (!split_pre || (pre && !split_mid) || (post && !split_post)) { + ret = -ENOMEM; + goto out; + } + + ASSERT(pre + post < len); + + lock_extent(&inode->io_tree, start, start + len - 1); + write_lock(&em_tree->lock); + em = lookup_extent_mapping(em_tree, start, len); + if (!em) { + ret = -EIO; + goto out_unlock; + } + + ASSERT(em->len == len); + ASSERT(!test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)); + ASSERT(em->block_start < EXTENT_MAP_LAST_BYTE); + + flags = em->flags; + clear_bit(EXTENT_FLAG_PINNED, &em->flags); + clear_bit(EXTENT_FLAG_LOGGING, &flags); + modified = !list_empty(&em->list); + + /* First, replace the em with a new extent_map starting from * em->start */ + split_pre->start = em->start; + split_pre->len = (pre ? pre : em->len - post); + split_pre->orig_start = split_pre->start; + split_pre->block_start = em->block_start; + split_pre->block_len = split_pre->len; + split_pre->orig_block_len = split_pre->block_len; + split_pre->ram_bytes = split_pre->len; + split_pre->flags = flags; + split_pre->compress_type = em->compress_type; + split_pre->generation = em->generation; + + replace_extent_mapping(em_tree, em, split_pre, modified); + + /* + * Now we only have an extent_map at: + * [em->start, em->start + pre] if pre != 0 + * [em->start, em->start + em->len - post] if pre == 0 + */ + + if (pre) { + /* Insert the middle extent_map */ + split_mid->start = em->start + pre; + split_mid->len = em->len - pre - post; + split_mid->orig_start = split_mid->start; + split_mid->block_start = em->block_start + pre; + split_mid->block_len = split_mid->len; + split_mid->orig_block_len = split_mid->block_len; + split_mid->ram_bytes = split_mid->len; + split_mid->flags = flags; + split_mid->compress_type = em->compress_type; + split_mid->generation = em->generation; + add_extent_mapping(em_tree, split_mid, modified); + } + + if (post) { + split_post->start = em->start + em->len - post; + split_post->len = post; + split_post->orig_start = split_post->start; + split_post->block_start = em->block_start + em->len - post; + split_post->block_len = split_post->len; + split_post->orig_block_len = split_post->block_len; + split_post->ram_bytes = split_post->len; + split_post->flags = flags; + split_post->compress_type = em->compress_type; + split_post->generation = em->generation; + add_extent_mapping(em_tree, split_post, modified); + } + + /* Once for us */ + free_extent_map(em); + /* Once for the tree */ + free_extent_map(em); + +out_unlock: + write_unlock(&em_tree->lock); + unlock_extent(&inode->io_tree, start, start + len - 1); +out: + free_extent_map(split_pre); + free_extent_map(split_mid); + free_extent_map(split_post); + + return ret; +} + static blk_status_t extract_ordered_extent(struct btrfs_inode *inode, struct bio *bio, loff_t file_offset) { struct btrfs_ordered_extent *ordered; - struct extent_map *em = NULL, *em_new = NULL; - struct extent_map_tree *em_tree = &inode->extent_tree; u64 start = (u64)bio->bi_iter.bi_sector << SECTOR_SHIFT; + u64 file_len; u64 len = bio->bi_iter.bi_size; u64 end = start + len; u64 ordered_end; @@ -2317,41 +2431,16 @@ static blk_status_t extract_ordered_extent(struct btrfs_inode *inode, goto out; } + file_len = ordered->num_bytes; pre = start - ordered->disk_bytenr; post = ordered_end - end; ret = btrfs_split_ordered_extent(ordered, pre, post); if (ret) goto out; - - read_lock(&em_tree->lock); - em = lookup_extent_mapping(em_tree, ordered->file_offset, len); - if (!em) { - read_unlock(&em_tree->lock); - ret = -EIO; - goto out; - } - read_unlock(&em_tree->lock); - - ASSERT(!test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)); - /* - * We cannot reuse em_new here but have to create a new one, as - * unpin_extent_cache() expects the start of the extent map to be the - * logical offset of the file, which does not hold true anymore after - * splitting. - */ - em_new = create_io_em(inode, em->start + pre, len, - em->start + pre, em->block_start + pre, len, - len, len, BTRFS_COMPRESS_NONE, - BTRFS_ORDERED_REGULAR); - if (IS_ERR(em_new)) { - ret = PTR_ERR(em_new); - goto out; - } - free_extent_map(em_new); + ret = split_zoned_em(inode, file_offset, file_len, pre, post); out: - free_extent_map(em); btrfs_put_ordered_extent(ordered); return errno_to_blk_status(ret); @@ -2903,7 +2992,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) goto out; } - if (ordered_extent->disk) + if (ordered_extent->bdev) btrfs_rewrite_logical_zoned(ordered_extent); btrfs_free_io_failure_record(inode, start, end); @@ -9137,8 +9226,14 @@ static int btrfs_rename_exchange(struct inode *old_dir, bool dest_log_pinned = false; bool need_abort = false; - /* we only allow rename subvolume link between subvolumes */ - if (old_ino != BTRFS_FIRST_FREE_OBJECTID && root != dest) + /* + * For non-subvolumes allow exchange only within one subvolume, in the + * same inode namespace. Two subvolumes (represented as directory) can + * be exchanged as they're a logical link and have a fixed inode number. + */ + if (root != dest && + (old_ino != BTRFS_FIRST_FREE_OBJECTID || + new_ino != BTRFS_FIRST_FREE_OBJECTID)) return -EXDEV; /* close the race window with snapshot create/destroy ioctl */ diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index 6eb41b7c0c84..5c0f8481e25e 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -190,8 +190,6 @@ static int __btrfs_add_ordered_extent(struct btrfs_inode *inode, u64 file_offset entry->truncated_len = (u64)-1; entry->qgroup_rsv = ret; entry->physical = (u64)-1; - entry->disk = NULL; - entry->partno = (u8)-1; ASSERT(type == BTRFS_ORDERED_REGULAR || type == BTRFS_ORDERED_NOCOW || diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h index 566472004edd..b2d88aba8420 100644 --- a/fs/btrfs/ordered-data.h +++ b/fs/btrfs/ordered-data.h @@ -145,8 +145,7 @@ struct btrfs_ordered_extent { * command in a workqueue context */ u64 physical; - struct gendisk *disk; - u8 partno; + struct block_device *bdev; }; /* diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 07ec06d4e972..0fa121171ca1 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1704,17 +1704,39 @@ int btrfs_qgroup_trace_extent_nolock(struct btrfs_fs_info *fs_info, return 0; } -int btrfs_qgroup_trace_extent_post(struct btrfs_fs_info *fs_info, +int btrfs_qgroup_trace_extent_post(struct btrfs_trans_handle *trans, struct btrfs_qgroup_extent_record *qrecord) { struct ulist *old_root; u64 bytenr = qrecord->bytenr; int ret; - ret = btrfs_find_all_roots(NULL, fs_info, bytenr, 0, &old_root, false); + /* + * We are always called in a context where we are already holding a + * transaction handle. Often we are called when adding a data delayed + * reference from btrfs_truncate_inode_items() (truncating or unlinking), + * in which case we will be holding a write lock on extent buffer from a + * subvolume tree. In this case we can't allow btrfs_find_all_roots() to + * acquire fs_info->commit_root_sem, because that is a higher level lock + * that must be acquired before locking any extent buffers. + * + * So we want btrfs_find_all_roots() to not acquire the commit_root_sem + * but we can't pass it a non-NULL transaction handle, because otherwise + * it would not use commit roots and would lock extent buffers, causing + * a deadlock if it ends up trying to read lock the same extent buffer + * that was previously write locked at btrfs_truncate_inode_items(). + * + * So pass a NULL transaction handle to btrfs_find_all_roots() and + * explicitly tell it to not acquire the commit_root_sem - if we are + * holding a transaction handle we don't need its protection. + */ + ASSERT(trans != NULL); + + ret = btrfs_find_all_roots(NULL, trans->fs_info, bytenr, 0, &old_root, + false, true); if (ret < 0) { - fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT; - btrfs_warn(fs_info, + trans->fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT; + btrfs_warn(trans->fs_info, "error accounting new delayed refs extent (err code: %d), quota inconsistent", ret); return 0; @@ -1758,7 +1780,7 @@ int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans, u64 bytenr, kfree(record); return 0; } - return btrfs_qgroup_trace_extent_post(fs_info, record); + return btrfs_qgroup_trace_extent_post(trans, record); } int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans, @@ -2629,7 +2651,7 @@ int btrfs_qgroup_account_extents(struct btrfs_trans_handle *trans) /* Search commit root to find old_roots */ ret = btrfs_find_all_roots(NULL, fs_info, record->bytenr, 0, - &record->old_roots, false); + &record->old_roots, false, false); if (ret < 0) goto cleanup; } @@ -2645,7 +2667,7 @@ int btrfs_qgroup_account_extents(struct btrfs_trans_handle *trans) * current root. It's safe inside commit_transaction(). */ ret = btrfs_find_all_roots(trans, fs_info, - record->bytenr, BTRFS_SEQ_LAST, &new_roots, false); + record->bytenr, BTRFS_SEQ_LAST, &new_roots, false, false); if (ret < 0) goto cleanup; if (qgroup_to_skip) { @@ -3179,7 +3201,7 @@ static int qgroup_rescan_leaf(struct btrfs_trans_handle *trans, num_bytes = found.offset; ret = btrfs_find_all_roots(NULL, fs_info, found.objectid, 0, - &roots, false); + &roots, false, false); if (ret < 0) goto out; /* For rescan, just pass old_roots as NULL */ diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h index 7283e4f549af..880e9df0dac1 100644 --- a/fs/btrfs/qgroup.h +++ b/fs/btrfs/qgroup.h @@ -298,7 +298,7 @@ int btrfs_qgroup_trace_extent_nolock( * using current root, then we can move all expensive backref walk out of * transaction committing, but not now as qgroup accounting will be wrong again. */ -int btrfs_qgroup_trace_extent_post(struct btrfs_fs_info *fs_info, +int btrfs_qgroup_trace_extent_post(struct btrfs_trans_handle *trans, struct btrfs_qgroup_extent_record *qrecord); /* diff --git a/fs/btrfs/tests/qgroup-tests.c b/fs/btrfs/tests/qgroup-tests.c index f3137285a9e2..98b5aaba46f1 100644 --- a/fs/btrfs/tests/qgroup-tests.c +++ b/fs/btrfs/tests/qgroup-tests.c @@ -224,7 +224,7 @@ static int test_no_shared_qgroup(struct btrfs_root *root, * quota. */ ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots, - false); + false, false); if (ret) { ulist_free(old_roots); test_err("couldn't find old roots: %d", ret); @@ -237,7 +237,7 @@ static int test_no_shared_qgroup(struct btrfs_root *root, return ret; ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots, - false); + false, false); if (ret) { ulist_free(old_roots); ulist_free(new_roots); @@ -261,7 +261,7 @@ static int test_no_shared_qgroup(struct btrfs_root *root, new_roots = NULL; ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots, - false); + false, false); if (ret) { ulist_free(old_roots); test_err("couldn't find old roots: %d", ret); @@ -273,7 +273,7 @@ static int test_no_shared_qgroup(struct btrfs_root *root, return -EINVAL; ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots, - false); + false, false); if (ret) { ulist_free(old_roots); ulist_free(new_roots); @@ -325,7 +325,7 @@ static int test_multiple_refs(struct btrfs_root *root, } ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots, - false); + false, false); if (ret) { ulist_free(old_roots); test_err("couldn't find old roots: %d", ret); @@ -338,7 +338,7 @@ static int test_multiple_refs(struct btrfs_root *root, return ret; ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots, - false); + false, false); if (ret) { ulist_free(old_roots); ulist_free(new_roots); @@ -360,7 +360,7 @@ static int test_multiple_refs(struct btrfs_root *root, } ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots, - false); + false, false); if (ret) { ulist_free(old_roots); test_err("couldn't find old roots: %d", ret); @@ -373,7 +373,7 @@ static int test_multiple_refs(struct btrfs_root *root, return ret; ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots, - false); + false, false); if (ret) { ulist_free(old_roots); ulist_free(new_roots); @@ -401,7 +401,7 @@ static int test_multiple_refs(struct btrfs_root *root, } ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots, - false); + false, false); if (ret) { ulist_free(old_roots); test_err("couldn't find old roots: %d", ret); @@ -414,7 +414,7 @@ static int test_multiple_refs(struct btrfs_root *root, return ret; ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots, - false); + false, false); if (ret) { ulist_free(old_roots); ulist_free(new_roots); diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 50318231c1a8..14b9fdc8aaa9 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -254,23 +254,21 @@ static inline int extwriter_counter_read(struct btrfs_transaction *trans) } /* - * To be called after all the new block groups attached to the transaction - * handle have been created (btrfs_create_pending_block_groups()). + * To be called after doing the chunk btree updates right after allocating a new + * chunk (after btrfs_chunk_alloc_add_chunk_item() is called), when removing a + * chunk after all chunk btree updates and after finishing the second phase of + * chunk allocation (btrfs_create_pending_block_groups()) in case some block + * group had its chunk item insertion delayed to the second phase. */ void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans) { struct btrfs_fs_info *fs_info = trans->fs_info; - struct btrfs_transaction *cur_trans = trans->transaction; if (!trans->chunk_bytes_reserved) return; - WARN_ON_ONCE(!list_empty(&trans->new_bgs)); - btrfs_block_rsv_release(fs_info, &fs_info->chunk_block_rsv, trans->chunk_bytes_reserved, NULL); - atomic64_sub(trans->chunk_bytes_reserved, &cur_trans->chunk_bytes_reserved); - cond_wake_up(&cur_trans->chunk_reserve_wait); trans->chunk_bytes_reserved = 0; } @@ -386,8 +384,6 @@ loop: spin_lock_init(&cur_trans->dropped_roots_lock); INIT_LIST_HEAD(&cur_trans->releasing_ebs); spin_lock_init(&cur_trans->releasing_ebs_lock); - atomic64_set(&cur_trans->chunk_bytes_reserved, 0); - init_waitqueue_head(&cur_trans->chunk_reserve_wait); list_add_tail(&cur_trans->list, &fs_info->trans_list); extent_io_tree_init(fs_info, &cur_trans->dirty_pages, IO_TREE_TRANS_DIRTY_PAGES, fs_info->btree_inode); @@ -701,7 +697,6 @@ again: h->fs_info = root->fs_info; h->type = type; - h->can_flush_pending_bgs = true; INIT_LIST_HEAD(&h->new_bgs); smp_mb(); diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index 07d76029f598..ba45065f9451 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -96,13 +96,6 @@ struct btrfs_transaction { spinlock_t releasing_ebs_lock; struct list_head releasing_ebs; - - /* - * The number of bytes currently reserved, by all transaction handles - * attached to this transaction, for metadata extents of the chunk tree. - */ - atomic64_t chunk_bytes_reserved; - wait_queue_head_t chunk_reserve_wait; }; #define __TRANS_FREEZABLE (1U << 0) @@ -139,7 +132,7 @@ struct btrfs_trans_handle { short aborted; bool adding_csums; bool allocating_chunk; - bool can_flush_pending_bgs; + bool removing_chunk; bool reloc_reserved; bool in_fsync; struct btrfs_root *root; diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index cab451d19547..e6430ac9bbe8 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -3173,7 +3173,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, if (!log_root_tree->node) { ret = btrfs_alloc_log_tree_node(trans, log_root_tree); if (ret) { - mutex_unlock(&fs_info->tree_log_mutex); + mutex_unlock(&fs_info->tree_root->log_mutex); goto out; } } @@ -5526,16 +5526,29 @@ log_extents: spin_lock(&inode->lock); inode->logged_trans = trans->transid; /* - * Don't update last_log_commit if we logged that an inode exists - * after it was loaded to memory (full_sync bit set). - * This is to prevent data loss when we do a write to the inode, - * then the inode gets evicted after all delalloc was flushed, - * then we log it exists (due to a rename for example) and then - * fsync it. This last fsync would do nothing (not logging the - * extents previously written). + * Don't update last_log_commit if we logged that an inode exists. + * We do this for two reasons: + * + * 1) We might have had buffered writes to this inode that were + * flushed and had their ordered extents completed in this + * transaction, but we did not previously log the inode with + * LOG_INODE_ALL. Later the inode was evicted and after that + * it was loaded again and this LOG_INODE_EXISTS log operation + * happened. We must make sure that if an explicit fsync against + * the inode is performed later, it logs the new extents, an + * updated inode item, etc, and syncs the log. The same logic + * applies to direct IO writes instead of buffered writes. + * + * 2) When we log the inode with LOG_INODE_EXISTS, its inode item + * is logged with an i_size of 0 or whatever value was logged + * before. If later the i_size of the inode is increased by a + * truncate operation, the log is synced through an fsync of + * some other inode and then finally an explicit fsync against + * this inode is made, we must make sure this fsync logs the + * inode with the new i_size, the hole between old i_size and + * the new i_size, and syncs the log. */ - if (inode_only != LOG_INODE_EXISTS || - !test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags)) + if (inode_only != LOG_INODE_EXISTS) inode->last_log_commit = inode->last_sub_trans; spin_unlock(&inode->lock); } @@ -6490,8 +6503,8 @@ void btrfs_log_new_name(struct btrfs_trans_handle *trans, * if this inode hasn't been logged and directory we're renaming it * from hasn't been logged, we don't need to log it */ - if (inode->logged_trans < trans->transid && - (!old_dir || old_dir->logged_trans < trans->transid)) + if (!inode_logged(trans, inode) && + (!old_dir || !inode_logged(trans, old_dir))) return; /* diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 807502cd6510..70f94b75f25a 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1078,6 +1078,7 @@ static void __btrfs_free_extra_devids(struct btrfs_fs_devices *fs_devices, if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) { list_del_init(&device->dev_alloc_list); clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state); + fs_devices->rw_devices--; } list_del_init(&device->dev_list); fs_devices->num_devices--; @@ -1745,19 +1746,14 @@ again: extent = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_dev_extent); } else { - btrfs_handle_fs_error(fs_info, ret, "Slot search failed"); goto out; } *dev_extent_len = btrfs_dev_extent_length(leaf, extent); ret = btrfs_del_item(trans, root, path); - if (ret) { - btrfs_handle_fs_error(fs_info, ret, - "Failed to remove dev extent item"); - } else { + if (ret == 0) set_bit(BTRFS_TRANS_HAVE_FREE_BGS, &trans->transaction->flags); - } out: btrfs_free_path(path); return ret; @@ -2942,7 +2938,7 @@ static int btrfs_del_sys_chunk(struct btrfs_fs_info *fs_info, u64 chunk_offset) u32 cur; struct btrfs_key key; - mutex_lock(&fs_info->chunk_mutex); + lockdep_assert_held(&fs_info->chunk_mutex); array_size = btrfs_super_sys_array_size(super_copy); ptr = super_copy->sys_chunk_array; @@ -2972,7 +2968,6 @@ static int btrfs_del_sys_chunk(struct btrfs_fs_info *fs_info, u64 chunk_offset) cur += len; } } - mutex_unlock(&fs_info->chunk_mutex); return ret; } @@ -3012,6 +3007,29 @@ struct extent_map *btrfs_get_chunk_map(struct btrfs_fs_info *fs_info, return em; } +static int remove_chunk_item(struct btrfs_trans_handle *trans, + struct map_lookup *map, u64 chunk_offset) +{ + int i; + + /* + * Removing chunk items and updating the device items in the chunks btree + * requires holding the chunk_mutex. + * See the comment at btrfs_chunk_alloc() for the details. + */ + lockdep_assert_held(&trans->fs_info->chunk_mutex); + + for (i = 0; i < map->num_stripes; i++) { + int ret; + + ret = btrfs_update_device(trans, map->stripes[i].dev); + if (ret) + return ret; + } + + return btrfs_free_chunk(trans, chunk_offset); +} + int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset) { struct btrfs_fs_info *fs_info = trans->fs_info; @@ -3032,14 +3050,16 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset) return PTR_ERR(em); } map = em->map_lookup; - mutex_lock(&fs_info->chunk_mutex); - check_system_chunk(trans, map->type); - mutex_unlock(&fs_info->chunk_mutex); /* - * Take the device list mutex to prevent races with the final phase of - * a device replace operation that replaces the device object associated - * with map stripes (dev-replace.c:btrfs_dev_replace_finishing()). + * First delete the device extent items from the devices btree. + * We take the device_list_mutex to avoid racing with the finishing phase + * of a device replace operation. See the comment below before acquiring + * fs_info->chunk_mutex. Note that here we do not acquire the chunk_mutex + * because that can result in a deadlock when deleting the device extent + * items from the devices btree - COWing an extent buffer from the btree + * may result in allocating a new metadata chunk, which would attempt to + * lock again fs_info->chunk_mutex. */ mutex_lock(&fs_devices->device_list_mutex); for (i = 0; i < map->num_stripes; i++) { @@ -3061,18 +3081,73 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset) btrfs_clear_space_info_full(fs_info); mutex_unlock(&fs_info->chunk_mutex); } + } + mutex_unlock(&fs_devices->device_list_mutex); - ret = btrfs_update_device(trans, device); + /* + * We acquire fs_info->chunk_mutex for 2 reasons: + * + * 1) Just like with the first phase of the chunk allocation, we must + * reserve system space, do all chunk btree updates and deletions, and + * update the system chunk array in the superblock while holding this + * mutex. This is for similar reasons as explained on the comment at + * the top of btrfs_chunk_alloc(); + * + * 2) Prevent races with the final phase of a device replace operation + * that replaces the device object associated with the map's stripes, + * because the device object's id can change at any time during that + * final phase of the device replace operation + * (dev-replace.c:btrfs_dev_replace_finishing()), so we could grab the + * replaced device and then see it with an ID of + * BTRFS_DEV_REPLACE_DEVID, which would cause a failure when updating + * the device item, which does not exists on the chunk btree. + * The finishing phase of device replace acquires both the + * device_list_mutex and the chunk_mutex, in that order, so we are + * safe by just acquiring the chunk_mutex. + */ + trans->removing_chunk = true; + mutex_lock(&fs_info->chunk_mutex); + + check_system_chunk(trans, map->type); + + ret = remove_chunk_item(trans, map, chunk_offset); + /* + * Normally we should not get -ENOSPC since we reserved space before + * through the call to check_system_chunk(). + * + * Despite our system space_info having enough free space, we may not + * be able to allocate extents from its block groups, because all have + * an incompatible profile, which will force us to allocate a new system + * block group with the right profile, or right after we called + * check_system_space() above, a scrub turned the only system block group + * with enough free space into RO mode. + * This is explained with more detail at do_chunk_alloc(). + * + * So if we get -ENOSPC, allocate a new system chunk and retry once. + */ + if (ret == -ENOSPC) { + const u64 sys_flags = btrfs_system_alloc_profile(fs_info); + struct btrfs_block_group *sys_bg; + + sys_bg = btrfs_alloc_chunk(trans, sys_flags); + if (IS_ERR(sys_bg)) { + ret = PTR_ERR(sys_bg); + btrfs_abort_transaction(trans, ret); + goto out; + } + + ret = btrfs_chunk_alloc_add_chunk_item(trans, sys_bg); if (ret) { - mutex_unlock(&fs_devices->device_list_mutex); btrfs_abort_transaction(trans, ret); goto out; } - } - mutex_unlock(&fs_devices->device_list_mutex); - ret = btrfs_free_chunk(trans, chunk_offset); - if (ret) { + ret = remove_chunk_item(trans, map, chunk_offset); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out; + } + } else if (ret) { btrfs_abort_transaction(trans, ret); goto out; } @@ -3087,6 +3162,15 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset) } } + mutex_unlock(&fs_info->chunk_mutex); + trans->removing_chunk = false; + + /* + * We are done with chunk btree updates and deletions, so release the + * system space we previously reserved (with check_system_chunk()). + */ + btrfs_trans_release_chunk_metadata(trans); + ret = btrfs_remove_block_group(trans, chunk_offset, em); if (ret) { btrfs_abort_transaction(trans, ret); @@ -3094,6 +3178,10 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset) } out: + if (trans->removing_chunk) { + mutex_unlock(&fs_info->chunk_mutex); + trans->removing_chunk = false; + } /* once for us */ free_extent_map(em); return ret; @@ -4860,13 +4948,12 @@ static int btrfs_add_system_chunk(struct btrfs_fs_info *fs_info, u32 array_size; u8 *ptr; - mutex_lock(&fs_info->chunk_mutex); + lockdep_assert_held(&fs_info->chunk_mutex); + array_size = btrfs_super_sys_array_size(super_copy); if (array_size + item_size + sizeof(disk_key) - > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) { - mutex_unlock(&fs_info->chunk_mutex); + > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) return -EFBIG; - } ptr = super_copy->sys_chunk_array + array_size; btrfs_cpu_key_to_disk(&disk_key, key); @@ -4875,7 +4962,6 @@ static int btrfs_add_system_chunk(struct btrfs_fs_info *fs_info, memcpy(ptr, chunk, item_size); item_size += sizeof(disk_key); btrfs_set_super_sys_array_size(super_copy, array_size + item_size); - mutex_unlock(&fs_info->chunk_mutex); return 0; } @@ -5225,13 +5311,14 @@ static int decide_stripe_size(struct btrfs_fs_devices *fs_devices, } } -static int create_chunk(struct btrfs_trans_handle *trans, +static struct btrfs_block_group *create_chunk(struct btrfs_trans_handle *trans, struct alloc_chunk_ctl *ctl, struct btrfs_device_info *devices_info) { struct btrfs_fs_info *info = trans->fs_info; struct map_lookup *map = NULL; struct extent_map_tree *em_tree; + struct btrfs_block_group *block_group; struct extent_map *em; u64 start = ctl->start; u64 type = ctl->type; @@ -5241,7 +5328,7 @@ static int create_chunk(struct btrfs_trans_handle *trans, map = kmalloc(map_lookup_size(ctl->num_stripes), GFP_NOFS); if (!map) - return -ENOMEM; + return ERR_PTR(-ENOMEM); map->num_stripes = ctl->num_stripes; for (i = 0; i < ctl->ndevs; ++i) { @@ -5263,7 +5350,7 @@ static int create_chunk(struct btrfs_trans_handle *trans, em = alloc_extent_map(); if (!em) { kfree(map); - return -ENOMEM; + return ERR_PTR(-ENOMEM); } set_bit(EXTENT_FLAG_FS_MAPPING, &em->flags); em->map_lookup = map; @@ -5279,12 +5366,12 @@ static int create_chunk(struct btrfs_trans_handle *trans, if (ret) { write_unlock(&em_tree->lock); free_extent_map(em); - return ret; + return ERR_PTR(ret); } write_unlock(&em_tree->lock); - ret = btrfs_make_block_group(trans, 0, type, start, ctl->chunk_size); - if (ret) + block_group = btrfs_make_block_group(trans, 0, type, start, ctl->chunk_size); + if (IS_ERR(block_group)) goto error_del_extent; for (i = 0; i < map->num_stripes; i++) { @@ -5304,7 +5391,7 @@ static int create_chunk(struct btrfs_trans_handle *trans, check_raid56_incompat_flag(info, type); check_raid1c34_incompat_flag(info, type); - return 0; + return block_group; error_del_extent: write_lock(&em_tree->lock); @@ -5316,34 +5403,36 @@ error_del_extent: /* One for the tree reference */ free_extent_map(em); - return ret; + return block_group; } -int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type) +struct btrfs_block_group *btrfs_alloc_chunk(struct btrfs_trans_handle *trans, + u64 type) { struct btrfs_fs_info *info = trans->fs_info; struct btrfs_fs_devices *fs_devices = info->fs_devices; struct btrfs_device_info *devices_info = NULL; struct alloc_chunk_ctl ctl; + struct btrfs_block_group *block_group; int ret; lockdep_assert_held(&info->chunk_mutex); if (!alloc_profile_is_valid(type, 0)) { ASSERT(0); - return -EINVAL; + return ERR_PTR(-EINVAL); } if (list_empty(&fs_devices->alloc_list)) { if (btrfs_test_opt(info, ENOSPC_DEBUG)) btrfs_debug(info, "%s: no writable device", __func__); - return -ENOSPC; + return ERR_PTR(-ENOSPC); } if (!(type & BTRFS_BLOCK_GROUP_TYPE_MASK)) { btrfs_err(info, "invalid chunk type 0x%llx requested", type); ASSERT(0); - return -EINVAL; + return ERR_PTR(-EINVAL); } ctl.start = find_next_chunk(info); @@ -5353,46 +5442,43 @@ int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type) devices_info = kcalloc(fs_devices->rw_devices, sizeof(*devices_info), GFP_NOFS); if (!devices_info) - return -ENOMEM; + return ERR_PTR(-ENOMEM); ret = gather_device_info(fs_devices, &ctl, devices_info); - if (ret < 0) + if (ret < 0) { + block_group = ERR_PTR(ret); goto out; + } ret = decide_stripe_size(fs_devices, &ctl, devices_info); - if (ret < 0) + if (ret < 0) { + block_group = ERR_PTR(ret); goto out; + } - ret = create_chunk(trans, &ctl, devices_info); + block_group = create_chunk(trans, &ctl, devices_info); out: kfree(devices_info); - return ret; + return block_group; } /* - * Chunk allocation falls into two parts. The first part does work - * that makes the new allocated chunk usable, but does not do any operation - * that modifies the chunk tree. The second part does the work that - * requires modifying the chunk tree. This division is important for the - * bootstrap process of adding storage to a seed btrfs. + * This function, btrfs_finish_chunk_alloc(), belongs to phase 2. + * + * See the comment at btrfs_chunk_alloc() for details about the chunk allocation + * phases. */ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans, u64 chunk_offset, u64 chunk_size) { struct btrfs_fs_info *fs_info = trans->fs_info; - struct btrfs_root *extent_root = fs_info->extent_root; - struct btrfs_root *chunk_root = fs_info->chunk_root; - struct btrfs_key key; struct btrfs_device *device; - struct btrfs_chunk *chunk; - struct btrfs_stripe *stripe; struct extent_map *em; struct map_lookup *map; - size_t item_size; u64 dev_offset; u64 stripe_size; - int i = 0; + int i; int ret = 0; em = btrfs_get_chunk_map(fs_info, chunk_offset, chunk_size); @@ -5400,53 +5486,117 @@ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans, return PTR_ERR(em); map = em->map_lookup; - item_size = btrfs_chunk_item_size(map->num_stripes); stripe_size = em->orig_block_len; - chunk = kzalloc(item_size, GFP_NOFS); - if (!chunk) { - ret = -ENOMEM; - goto out; - } - /* * Take the device list mutex to prevent races with the final phase of * a device replace operation that replaces the device object associated * with the map's stripes, because the device object's id can change * at any time during that final phase of the device replace operation - * (dev-replace.c:btrfs_dev_replace_finishing()). + * (dev-replace.c:btrfs_dev_replace_finishing()), so we could grab the + * replaced device and then see it with an ID of BTRFS_DEV_REPLACE_DEVID, + * resulting in persisting a device extent item with such ID. */ mutex_lock(&fs_info->fs_devices->device_list_mutex); for (i = 0; i < map->num_stripes; i++) { device = map->stripes[i].dev; dev_offset = map->stripes[i].physical; - ret = btrfs_update_device(trans, device); - if (ret) - break; ret = btrfs_alloc_dev_extent(trans, device, chunk_offset, dev_offset, stripe_size); if (ret) break; } - if (ret) { - mutex_unlock(&fs_info->fs_devices->device_list_mutex); + mutex_unlock(&fs_info->fs_devices->device_list_mutex); + + free_extent_map(em); + return ret; +} + +/* + * This function, btrfs_chunk_alloc_add_chunk_item(), typically belongs to the + * phase 1 of chunk allocation. It belongs to phase 2 only when allocating system + * chunks. + * + * See the comment at btrfs_chunk_alloc() for details about the chunk allocation + * phases. + */ +int btrfs_chunk_alloc_add_chunk_item(struct btrfs_trans_handle *trans, + struct btrfs_block_group *bg) +{ + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_root *extent_root = fs_info->extent_root; + struct btrfs_root *chunk_root = fs_info->chunk_root; + struct btrfs_key key; + struct btrfs_chunk *chunk; + struct btrfs_stripe *stripe; + struct extent_map *em; + struct map_lookup *map; + size_t item_size; + int i; + int ret; + + /* + * We take the chunk_mutex for 2 reasons: + * + * 1) Updates and insertions in the chunk btree must be done while holding + * the chunk_mutex, as well as updating the system chunk array in the + * superblock. See the comment on top of btrfs_chunk_alloc() for the + * details; + * + * 2) To prevent races with the final phase of a device replace operation + * that replaces the device object associated with the map's stripes, + * because the device object's id can change at any time during that + * final phase of the device replace operation + * (dev-replace.c:btrfs_dev_replace_finishing()), so we could grab the + * replaced device and then see it with an ID of BTRFS_DEV_REPLACE_DEVID, + * which would cause a failure when updating the device item, which does + * not exists, or persisting a stripe of the chunk item with such ID. + * Here we can't use the device_list_mutex because our caller already + * has locked the chunk_mutex, and the final phase of device replace + * acquires both mutexes - first the device_list_mutex and then the + * chunk_mutex. Using any of those two mutexes protects us from a + * concurrent device replace. + */ + lockdep_assert_held(&fs_info->chunk_mutex); + + em = btrfs_get_chunk_map(fs_info, bg->start, bg->length); + if (IS_ERR(em)) { + ret = PTR_ERR(em); + btrfs_abort_transaction(trans, ret); + return ret; + } + + map = em->map_lookup; + item_size = btrfs_chunk_item_size(map->num_stripes); + + chunk = kzalloc(item_size, GFP_NOFS); + if (!chunk) { + ret = -ENOMEM; + btrfs_abort_transaction(trans, ret); goto out; } + for (i = 0; i < map->num_stripes; i++) { + struct btrfs_device *device = map->stripes[i].dev; + + ret = btrfs_update_device(trans, device); + if (ret) + goto out; + } + stripe = &chunk->stripe; for (i = 0; i < map->num_stripes; i++) { - device = map->stripes[i].dev; - dev_offset = map->stripes[i].physical; + struct btrfs_device *device = map->stripes[i].dev; + const u64 dev_offset = map->stripes[i].physical; btrfs_set_stack_stripe_devid(stripe, device->devid); btrfs_set_stack_stripe_offset(stripe, dev_offset); memcpy(stripe->dev_uuid, device->uuid, BTRFS_UUID_SIZE); stripe++; } - mutex_unlock(&fs_info->fs_devices->device_list_mutex); - btrfs_set_stack_chunk_length(chunk, chunk_size); + btrfs_set_stack_chunk_length(chunk, bg->length); btrfs_set_stack_chunk_owner(chunk, extent_root->root_key.objectid); btrfs_set_stack_chunk_stripe_len(chunk, map->stripe_len); btrfs_set_stack_chunk_type(chunk, map->type); @@ -5458,15 +5608,18 @@ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans, key.objectid = BTRFS_FIRST_CHUNK_TREE_OBJECTID; key.type = BTRFS_CHUNK_ITEM_KEY; - key.offset = chunk_offset; + key.offset = bg->start; ret = btrfs_insert_item(trans, chunk_root, &key, chunk, item_size); - if (ret == 0 && map->type & BTRFS_BLOCK_GROUP_SYSTEM) { - /* - * TODO: Cleanup of inserted chunk root in case of - * failure. - */ + if (ret) + goto out; + + bg->chunk_item_inserted = 1; + + if (map->type & BTRFS_BLOCK_GROUP_SYSTEM) { ret = btrfs_add_system_chunk(fs_info, &key, chunk, item_size); + if (ret) + goto out; } out: @@ -5479,16 +5632,41 @@ static noinline int init_first_rw_device(struct btrfs_trans_handle *trans) { struct btrfs_fs_info *fs_info = trans->fs_info; u64 alloc_profile; - int ret; + struct btrfs_block_group *meta_bg; + struct btrfs_block_group *sys_bg; + + /* + * When adding a new device for sprouting, the seed device is read-only + * so we must first allocate a metadata and a system chunk. But before + * adding the block group items to the extent, device and chunk btrees, + * we must first: + * + * 1) Create both chunks without doing any changes to the btrees, as + * otherwise we would get -ENOSPC since the block groups from the + * seed device are read-only; + * + * 2) Add the device item for the new sprout device - finishing the setup + * of a new block group requires updating the device item in the chunk + * btree, so it must exist when we attempt to do it. The previous step + * ensures this does not fail with -ENOSPC. + * + * After that we can add the block group items to their btrees: + * update existing device item in the chunk btree, add a new block group + * item to the extent btree, add a new chunk item to the chunk btree and + * finally add the new device extent items to the devices btree. + */ alloc_profile = btrfs_metadata_alloc_profile(fs_info); - ret = btrfs_alloc_chunk(trans, alloc_profile); - if (ret) - return ret; + meta_bg = btrfs_alloc_chunk(trans, alloc_profile); + if (IS_ERR(meta_bg)) + return PTR_ERR(meta_bg); alloc_profile = btrfs_system_alloc_profile(fs_info); - ret = btrfs_alloc_chunk(trans, alloc_profile); - return ret; + sys_bg = btrfs_alloc_chunk(trans, alloc_profile); + if (IS_ERR(sys_bg)) + return PTR_ERR(sys_bg); + + return 0; } static inline int btrfs_chunk_max_errors(struct map_lookup *map) @@ -7415,10 +7593,18 @@ int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info) total_dev++; } else if (found_key.type == BTRFS_CHUNK_ITEM_KEY) { struct btrfs_chunk *chunk; + + /* + * We are only called at mount time, so no need to take + * fs_info->chunk_mutex. Plus, to avoid lockdep warnings, + * we always lock first fs_info->chunk_mutex before + * acquiring any locks on the chunk tree. This is a + * requirement for chunk allocation, see the comment on + * top of btrfs_chunk_alloc() for details. + */ + ASSERT(!test_bit(BTRFS_FS_OPEN, &fs_info->flags)); chunk = btrfs_item_ptr(leaf, slot, struct btrfs_chunk); - mutex_lock(&fs_info->chunk_mutex); ret = read_one_chunk(&found_key, leaf, chunk); - mutex_unlock(&fs_info->chunk_mutex); if (ret) goto error; } diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index c7fc7caf575c..55a8ba244716 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -450,7 +450,8 @@ int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, struct extent_map *map, struct btrfs_io_geometry *io_geom); int btrfs_read_sys_array(struct btrfs_fs_info *fs_info); int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info); -int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type); +struct btrfs_block_group *btrfs_alloc_chunk(struct btrfs_trans_handle *trans, + u64 type); void btrfs_mapping_tree_free(struct extent_map_tree *tree); blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, int mirror_num); @@ -509,6 +510,8 @@ unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info, u64 logical); int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans, u64 chunk_offset, u64 chunk_size); +int btrfs_chunk_alloc_add_chunk_item(struct btrfs_trans_handle *trans, + struct btrfs_block_group *bg); int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset); struct extent_map *btrfs_get_chunk_map(struct btrfs_fs_info *fs_info, u64 logical, u64 length); diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index 297c0b1c0634..907c2cc45c9c 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -1349,8 +1349,7 @@ void btrfs_record_physical_zoned(struct inode *inode, u64 file_offset, return; ordered->physical = physical; - ordered->disk = bio->bi_bdev->bd_disk; - ordered->partno = bio->bi_bdev->bd_partno; + ordered->bdev = bio->bi_bdev; btrfs_put_ordered_extent(ordered); } @@ -1362,18 +1361,16 @@ void btrfs_rewrite_logical_zoned(struct btrfs_ordered_extent *ordered) struct extent_map_tree *em_tree; struct extent_map *em; struct btrfs_ordered_sum *sum; - struct block_device *bdev; u64 orig_logical = ordered->disk_bytenr; u64 *logical = NULL; int nr, stripe_len; /* Zoned devices should not have partitions. So, we can assume it is 0 */ - ASSERT(ordered->partno == 0); - bdev = bdgrab(ordered->disk->part0); - if (WARN_ON(!bdev)) + ASSERT(!bdev_is_partition(ordered->bdev)); + if (WARN_ON(!ordered->bdev)) return; - if (WARN_ON(btrfs_rmap_block(fs_info, orig_logical, bdev, + if (WARN_ON(btrfs_rmap_block(fs_info, orig_logical, ordered->bdev, ordered->physical, &logical, &nr, &stripe_len))) goto out; @@ -1402,7 +1399,6 @@ void btrfs_rewrite_logical_zoned(struct btrfs_ordered_extent *ordered) out: kfree(logical); - bdput(bdev); } bool btrfs_check_meta_write_pointer(struct btrfs_fs_info *fs_info, |